A new study finds that you may be able to spot a certain type of cancer early on—by looking at pictures of your baby. The research was Bryan Shaw's idea: In addition to being a professor of chemistry, Shaw is also a dad whose son, Noah, was diagnosed with an eye cancer known as retinoblastoma when he was just months old, Popular Science reports. But months before Noah's diagnosis, Shaw's wife noticed that in some pictures (including photos from when Noah was as young as 12 days old), Noah experienced "red eye"—but one of his eyes looked milky-white, not red. After Noah was diagnosed with the rare disease (he lost the eye, but is now healthy), his dad teamed up with the little boy's doctors to start the research—and they found that the milkiness in Noah's photos was no fluke.
They looked at more than 7,000 photos from the Shaws and other families, and found that digital pictures can, in fact, reveal leukocoria, the syndrome that led to Noah's retinoblastoma diagnosis—even when a camera's red-eye reduction setting is used. The milkiness occurs because tumors are developing in the back of the eye, and light reflects off them. They also found that the more milkiness the photos reveal, the bigger the tumor, NPR reports. The research could be particularly helpful if a cheap screening tool can be developed for developing countries: Though the survival rate for retinoblastoma is 95% in the US, it's less than 46% in Namibia. In areas like that, "People have a camera phone even if they don’t have a washer or a car," Shaw says. "In places like China or India, peoples’ access to digital photography increases at a faster rate than their access to doctors does." (Another potential medical breakthrough for kids: A boy's peanut allergy was apparently cured after he got a bone marrow transplant.)